11 January 2017
This week we learned about the important role that Geography plays in spatial analysis. The following are some major themes that were discussed in class:
Scale
The parameters of interest in your study are a function of the scale at which you approach the problem. For example, if you are interested in a 25m x 25m patch of trees, it is not necessary to consider soil variability because the soil type is uniform throughout the study area. This however is not true if you are studying an entire valley for example. In this case you would definitely have to consider the various different soils that occur throughout the valley.
Scale can also affect spatial autocorrelation (SA). The type and magnitude of correlation may change as you increase or decrease the extent of your study. For example, it is possible that negative SA is observed within a patch of sea grass and positive SA is observed between patches of sea grass.
As such, multi-scale analyses are often needed in order to understand all of the processes at work.
Modifiable Areal Unit Problem (MAUP)
The MAUP is made up of two components: the scale effect and aggregation problem and the uncertainty in choosing zonal units. The scale effect refers to the fact that different statistical results are obtained from the same set of data when grouped at different scales and the aggregation problem refers to the fact that different statistical results can be obtained when data are grouped in different ways at the same scale. The uncertainty of choosing zonal units refers to the fact that “different areal arrangements of the same data produce different results”. This can be seen in Gerrymandering.
The MAUP exists because similar species/ phenomena tend to occur close together and are not grouped randomly. There are three Neighbourhood models that explain the relationship between non-random groups:
- Grouping – there is some kind of filtering mechanism at play which constrains people to locate in a specific area. Eg. in the past, the Chinese were only permitted to live in Chinatown .
- Group-dependent – “individuals in the same area are subject to similar external influences”. Eg. The West Vancouver Covenant restricted West Vancouver home owners from selling their homes to non-whites.
- Feedback – the frequency and strength of the interaction between the individuals within a group is greater than that with members outside of the group. Eg. Bog plants maintains the acidity of the bog and makes it difficult for individuals who are not accustomed to the acidic environment to enter.
Keywords:
Characteristic scale – a scale in which the phenomenon of interest naturally occurs. For example, if you are conducting a study on tidal animals, the spatial scale of your study is most likely within the tidal zone (and would not extend into the deep ocean where tidal animals do not occur).
Ecological fallacy – when an individual is assumed to possess the characteristics of the group to which s/he belongs. For example, assuming that all individuals in a given CT makes $30, 000/ year because the average income of the CT is reported to be $30, 000/year.
Individualistic fallacy – when a whole group is assumed to possess the characteristics of an individual that belongs to the group. For example, assuming that all UBC students are extremely good at physics because student A from UBC won a Nobel Prize in Physics.
References
Klinkenberg, B. (2017). Why is ‘geography’ important. [Lecture notes]. Retrieved from http://ibis.geog.ubc.ca/courses/geob479/notes/Handouts/Lecture02.pdf